
# Import all the libraries
import numpy as np # linear algebra
import pandas as pd # data processing, data manupilation

# visualization
import matplotlib.pyplot as plt
from matplotlib import rcParams

import seaborn as sns
%matplotlib inline
import plotly.offline as py
import plotly.graph_objs as go
import plotly.subplots as sp 
import plotly.figure_factory as ff

#remove warnings
import warnings
warnings.filterwarnings("ignore")


data = pd.read_csv("bank-customer-churn.csv", index_col=0)


data


data.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 10000 entries, 0 to 9999
Data columns (total 14 columns):
 #   Column            Non-Null Count  Dtype  
---  ------            --------------  -----  
 0   Row Number        10000 non-null  int64  
 1   Customer Id       10000 non-null  int64  
 2   Surname           10000 non-null  object 
 3   Credit Score      10000 non-null  object 
 4   Geography         10000 non-null  object 
 5   Gender            10000 non-null  object 
 6   Age               10000 non-null  int64  
 7   Tenure            10000 non-null  int64  
 8   Balance           10000 non-null  float64
 9   Num Of Products   10000 non-null  int64  
 10  Has CrCard        10000 non-null  object 
 11  Is Active Member  10000 non-null  object 
 12  Estimated Salary  10000 non-null  object 
 13  Exited            10000 non-null  object 
dtypes: float64(1), int64(5), object(8)
memory usage: 1.1+ MB


data.describe()


print ("Rows     : " ,data.shape[0])
print ("Columns  : " ,data.shape[1])
print ("\nFeatures : \n" ,data.columns.tolist())
print ("\nUnique values :  \n",data.nunique())

Rows     :  10000
Columns  :  14

Features : 
 ['Row Number', 'Customer Id', 'Surname', 'Credit Score', 'Geography', 'Gender', 'Age', 'Tenure', 'Balance', 'Num Of Products', 'Has CrCard', 'Is Active Member', 'Estimated Salary', 'Exited']

Unique values :  
 Row Number          10000
Customer Id         10000
Surname              2932
Credit Score          461
Geography               3
Gender                  2
Age                    70
Tenure                 11
Balance              6382
Num Of Products         4
Has CrCard              2
Is Active Member        2
Estimated Salary     9993
Exited                  2
dtype: int64


data.isnull().sum()

Row Number          0
Customer Id         0
Surname             0
Credit Score        0
Geography           0
Gender              0
Age                 0
Tenure              0
Balance             0
Num Of Products     0
Has CrCard          0
Is Active Member    0
Estimated Salary    0
Exited              0
dtype: int64


headers = ['RowNumber', 'CustomerId', 'Surname', 'CreditScore', 'Geography','Gender','Age','Tenure',
           'Balance','NumOfProducts','HasCrCard','IsActiveMember','EstimatedSalary', 'Churn']


df1= data.copy()


df1.columns = headers


df1.head()


data.head()


missing_data = df1.isnull()


# Count missing values
for column in missing_data.columns.values.tolist():
    print(column)
    print (missing_data[column].value_counts())
    print("")

RowNumber
False    10000
Name: RowNumber, dtype: int64

CustomerId
False    10000
Name: CustomerId, dtype: int64

Surname
False    10000
Name: Surname, dtype: int64

CreditScore
False    10000
Name: CreditScore, dtype: int64

Geography
False    10000
Name: Geography, dtype: int64

Gender
False    10000
Name: Gender, dtype: int64

Age
False    10000
Name: Age, dtype: int64

Tenure
False    10000
Name: Tenure, dtype: int64

Balance
False    10000
Name: Balance, dtype: int64

NumOfProducts
False    10000
Name: NumOfProducts, dtype: int64

HasCrCard
False    10000
Name: HasCrCard, dtype: int64

IsActiveMember
False    10000
Name: IsActiveMember, dtype: int64

EstimatedSalary
False    10000
Name: EstimatedSalary, dtype: int64

Churn
False    10000
Name: Churn, dtype: int64


#Replacing spaces with null values
df1 = df1.replace(" ",np.nan)


missing_data = df1.isnull()


#use code for count missing values


#Dropping null values from credit score column which contain 0.04% missing data 
df1 = df1[df1["CreditScore"].notnull()]
df1 = df1.reset_index()[df1.columns]

#convert to float type
df1["CreditScore"] = df1["CreditScore"].astype(float)


df1.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9996 entries, 0 to 9995
Data columns (total 14 columns):
 #   Column           Non-Null Count  Dtype  
---  ------           --------------  -----  
 0   RowNumber        9996 non-null   int64  
 1   CustomerId       9996 non-null   int64  
 2   Surname          9996 non-null   object 
 3   CreditScore      9996 non-null   float64
 4   Geography        9996 non-null   object 
 5   Gender           9996 non-null   object 
 6   Age              9996 non-null   int64  
 7   Tenure           9996 non-null   int64  
 8   Balance          9996 non-null   float64
 9   NumOfProducts    9996 non-null   int64  
 10  HasCrCard        9996 non-null   object 
 11  IsActiveMember   9996 non-null   object 
 12  EstimatedSalary  9991 non-null   object 
 13  Churn            9996 non-null   object 
dtypes: float64(2), int64(5), object(7)
memory usage: 1.1+ MB


# Write code for Dropping null values from Estimated Salary column which contain 0.04% missing data 


#convert to float type
df1["EstimatedSalary"] = df1["EstimatedSalary"].astype(float)


df1.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9996 entries, 0 to 9995
Data columns (total 14 columns):
 #   Column           Non-Null Count  Dtype  
---  ------           --------------  -----  
 0   RowNumber        9996 non-null   int64  
 1   CustomerId       9996 non-null   int64  
 2   Surname          9996 non-null   object 
 3   CreditScore      9996 non-null   float64
 4   Geography        9996 non-null   object 
 5   Gender           9996 non-null   object 
 6   Age              9996 non-null   int64  
 7   Tenure           9996 non-null   int64  
 8   Balance          9996 non-null   float64
 9   NumOfProducts    9996 non-null   int64  
 10  HasCrCard        9996 non-null   object 
 11  IsActiveMember   9996 non-null   object 
 12  EstimatedSalary  9991 non-null   float64
 13  Churn            9996 non-null   object 
dtypes: float64(3), int64(5), object(6)
memory usage: 1.1+ MB

df1


#get dummies for HasCrCard
dv1 = pd.get_dummies(df1["HasCrCard"])
dv1.head()


#Rename columns
dv1.rename(columns={'Yes':'HasCrCard_Yes', 'No':'HasCrCard_No'}, inplace=True)
dv1.head()


# merge data
df1 = pd.concat([df1, dv1], axis=1)


df1.head()


# drop original HasCrCard
df1.drop("HasCrCard", axis = 1, inplace=True)


df1.head()


# write code for get_dummies method for IsActiveMember


df1.Tenure.unique()

array([ 2,  1,  8,  7,  4,  6,  3, 10,  5,  9,  0], dtype=int64)


# Convert Tenure to categorical column
def tenure_lab(df1) :
    
    if df1["Tenure"] <= 3 :
        return "Tenure_0-3"
    elif (df1["Tenure"] > 3) & (df1["Tenure"] <= 6 ):
        return "Tenure_3-6"
    elif (df1["Tenure"] > 6) & (df1["Tenure"] <= 10) :
        return "Tenure_6-10"


df1["Tenure_group"] = df1.apply(lambda df1:tenure_lab(df1),
                                      axis = 1)


df1.Tenure_group.unique()

array(['Tenure_0-3', 'Tenure_6-10', 'Tenure_3-6'], dtype=object)


df1.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9996 entries, 0 to 9995
Data columns (total 16 columns):
 #   Column           Non-Null Count  Dtype  
---  ------           --------------  -----  
 0   RowNumber        9996 non-null   int64  
 1   CustomerId       9996 non-null   int64  
 2   Surname          9996 non-null   object 
 3   CreditScore      9996 non-null   float64
 4   Geography        9996 non-null   object 
 5   Gender           9996 non-null   object 
 6   Age              9996 non-null   int64  
 7   Tenure           9996 non-null   int64  
 8   Balance          9996 non-null   float64
 9   NumOfProducts    9996 non-null   int64  
 10  IsActiveMember   9996 non-null   object 
 11  EstimatedSalary  9991 non-null   float64
 12  Churn            9996 non-null   object 
 13  HasCrCard_No     9996 non-null   uint8  
 14  HasCrCard_Yes    9996 non-null   uint8  
 15  Tenure_group     9996 non-null   object 
dtypes: float64(3), int64(5), object(6), uint8(2)
memory usage: 1.1+ MB


#Separating churn and non churn customers

churn     = df1[df1["Churn"] == 'Yes']
not_churn = df1[df1["Churn"] == 'No']


print ('Customers who stayed', len(not_churn))
print ('Customers who left', len(churn))

Customers who stayed 7961
Customers who left 2035


# rcParams for upcoming plots
rcParams['figure.figsize'] = 8,6


# Create a scatter plot
sns.scatterplot(data = df1, x =df1['Tenure'], y = df1['Balance'], hue=df1.Churn.tolist())

<AxesSubplot:xlabel='Tenure', ylabel='Balance'>


sns.boxplot(data= df1, x="Churn", y="Balance")

<AxesSubplot:xlabel='Churn', ylabel='Balance'>


#labels
lab = df1["Churn"].value_counts().keys().tolist()
#values
val = df1["Churn"].value_counts().values.tolist()

trace = go.Pie(labels = lab ,
               values = val ,
               marker = dict(colors =  [ 'royalblue' ,'lime'],
                             line = dict(color = "white",
                                         width =  1.3)
                            ),
               rotation = 90,
               hoverinfo = "label+value+text",
               hole = .5
              )
layout = go.Layout(dict(title = "Churn vs. Non-Churn",
                        plot_bgcolor  = "rgb(243,243,243)",
                        paper_bgcolor = "rgb(243,243,243)",
                       )
                  )

data = [trace]
fig = go.Figure(data = data,layout = layout)

py.iplot(fig)


def plot_pie(column) :
    
    trace1 = go.Pie(values  = churn[column].value_counts().values.tolist(),
                    labels  = churn[column].value_counts().keys().tolist(),
                    hoverinfo = "label+percent+name",
                    domain  = dict(x = [0,.48]),
                    name    = "Churn Customers",
                    marker  = dict(line = dict(width = 2,
                                               color = "rgb(243,243,243)")
                                  ),
                    hole    = .6
                   )
    trace2 = go.Pie(values  = not_churn[column].value_counts().values.tolist(),
                    labels  = not_churn[column].value_counts().keys().tolist(),
                    hoverinfo = "label+percent+name",
                    marker  = dict(line = dict(width = 2,
                                               color = "rgb(243,243,243)")
                                  ),
                    domain  = dict(x = [.52,1]),
                    hole    = .6,
                    name    = "Non churn customers" 
                   )


    layout = go.Layout(dict(title = column + " distribution in customer Churn ",
                            plot_bgcolor  = "rgb(243,243,243)",
                            paper_bgcolor = "rgb(243,243,243)",
                            annotations = [dict(text = "churn customers",
                                                font = dict(size = 13),
                                                showarrow = False,
                                                x = .15, y = .5),
                                           dict(text = "Non churn customers",
                                                font = dict(size = 13),
                                                showarrow = False,
                                                x = .88,y = .5
                                               )
                                          ]
                           )
                      )
    data = [trace1,trace2]
    fig  = go.Figure(data = data,layout = layout)
    py.iplot(fig)


df1.columns

Index(['RowNumber', 'CustomerId', 'Surname', 'CreditScore', 'Geography',
       'Gender', 'Age', 'Tenure', 'Balance', 'NumOfProducts', 'IsActiveMember',
       'EstimatedSalary', 'Churn', 'HasCrCard_No', 'HasCrCard_Yes',
       'Tenure_group'],
      dtype='object')


cat_cols = ['Geography', 'Gender', 'Tenure_group','NumOfProducts']


# for all categorical columns plot pie
for i in cat_cols :
    plot_pie(i)


#function  for histogram for customer attrition types
def histogram(column) :
    trace1 = go.Histogram(x  = churn[column],
                          histnorm= "percent",
                          name = "Churn Customers",
                          marker = dict(line = dict(width = .5,
                                                    color = "black"
                                                    )
                                        ),
                         opacity = .9 
                         ) 
    
    trace2 = go.Histogram(x  = not_churn[column],
                          histnorm = "percent",
                          name = "Non churn customers",
                          marker = dict(line = dict(width = .5,
                                              color = "black"
                                             )
                                 ),
                          opacity = .9
                         )
    
    data = [trace1,trace2]
    layout = go.Layout(dict(title =column + " distribution in customer Churn ",
                            plot_bgcolor  = "rgb(243,243,243)",
                            paper_bgcolor = "rgb(243,243,243)",
                            xaxis = dict(gridcolor = 'rgb(255, 255, 255)',
                                             title = column,
                                             zerolinewidth=1,
                                             ticklen=5,
                                             gridwidth=2
                                            ),
                            yaxis = dict(gridcolor = 'rgb(255, 255, 255)',
                                             title = "percent",
                                             zerolinewidth=1,
                                             ticklen=5,
                                             gridwidth=2
                                            ),
                           )
                      )
    fig  = go.Figure(data=data,layout=layout)
    
    py.iplot(fig)


df1.columns

Index(['RowNumber', 'CustomerId', 'Surname', 'CreditScore', 'Geography',
       'Gender', 'Age', 'Tenure', 'Balance', 'NumOfProducts', 'IsActiveMember',
       'EstimatedSalary', 'Churn', 'HasCrCard_No', 'HasCrCard_Yes',
       'Tenure_group'],
      dtype='object')


num_cols = ['Age','CreditScore','Tenure', 'Balance','EstimatedSalary']


#for all categorical columns plot histogram    
for i in num_cols :
    histogram(i)
plt.show()


# Save the cleaned file
df1.to_csv('clean_df.csv')

	Row Number	Customer Id	Surname	Credit Score	Geography	Gender	Age	Tenure	Balance	Num Of Products	Has CrCard	Is Active Member	Estimated Salary	Exited
0	1	15634602	Hargrave	619	France	Female	42	2	0.00	1	Yes	Yes	101348.88	Yes
1	2	15647311	Hill	608	Spain	Female	41	1	83807.86	1	No	Yes	112542.58	No
2	3	15619304	Onio	502	France	Female	42	8	159660.80	3	Yes	No	113931.57	Yes
3	4	15701354	Boni	699	France	Female	39	1	0.00	2	No	No	93826.63	No
4	5	15737888	Mitchell	850	Spain	Female	43	2	125510.82	1	Yes	Yes	79084.1	No
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9995	9996	15606229	Obijiaku	771	France	Male	39	5	0.00	2	Yes	No	96270.64	No
9996	9997	15569892	Johnstone	516	France	Male	35	10	57369.61	1	Yes	Yes	101699.77	No
9997	9998	15584532	Liu	709	France	Female	36	7	0.00	1	No	Yes	42085.58	Yes
9998	9999	15682355	Sabbatini	772	Germany	Male	42	3	75075.31	2	Yes	No	92888.52	Yes
9999	10000	15628319	Walker	792	France	Female	28	4	130142.79	1	Yes	No	38190.78	No

	Row Number	Customer Id	Age	Tenure	Balance	Num Of Products
count	10000.00000	1.000000e+04	10000.000000	10000.000000	10000.000000	10000.000000
mean	5000.50000	1.569094e+07	38.921800	5.012800	76485.889288	1.530200
std	2886.89568	7.193619e+04	10.487806	2.892174	62397.405202	0.581654
min	1.00000	1.556570e+07	18.000000	0.000000	0.000000	1.000000
25%	2500.75000	1.562853e+07	32.000000	3.000000	0.000000	1.000000
50%	5000.50000	1.569074e+07	37.000000	5.000000	97198.540000	1.000000
75%	7500.25000	1.575323e+07	44.000000	7.000000	127644.240000	2.000000
max	10000.00000	1.581569e+07	92.000000	10.000000	250898.090000	4.000000

	RowNumber	CustomerId	Surname	CreditScore	Geography	Gender	Age	Tenure	Balance	NumOfProducts	HasCrCard	IsActiveMember	EstimatedSalary	Churn
0	1	15634602	Hargrave	619	France	Female	42	2	0.00	1	Yes	Yes	101348.88	Yes
1	2	15647311	Hill	608	Spain	Female	41	1	83807.86	1	No	Yes	112542.58	No
2	3	15619304	Onio	502	France	Female	42	8	159660.80	3	Yes	No	113931.57	Yes
3	4	15701354	Boni	699	France	Female	39	1	0.00	2	No	No	93826.63	No
4	5	15737888	Mitchell	850	Spain	Female	43	2	125510.82	1	Yes	Yes	79084.1	No

	Row Number	Customer Id	Surname	Credit Score	Geography	Gender	Age	Tenure	Balance	Num Of Products	Has CrCard	Is Active Member	Estimated Salary	Exited
0	1	15634602	Hargrave	619	France	Female	42	2	0.00	1	Yes	Yes	101348.88	Yes
1	2	15647311	Hill	608	Spain	Female	41	1	83807.86	1	No	Yes	112542.58	No
2	3	15619304	Onio	502	France	Female	42	8	159660.80	3	Yes	No	113931.57	Yes
3	4	15701354	Boni	699	France	Female	39	1	0.00	2	No	No	93826.63	No
4	5	15737888	Mitchell	850	Spain	Female	43	2	125510.82	1	Yes	Yes	79084.1	No

	RowNumber	CustomerId	Surname	CreditScore	Geography	Gender	Age	Tenure	Balance	NumOfProducts	HasCrCard	IsActiveMember	EstimatedSalary	Churn
0	1	15634602	Hargrave	619.0	France	Female	42	2	0.00	1	Yes	Yes	101348.88	Yes
1	2	15647311	Hill	608.0	Spain	Female	41	1	83807.86	1	No	Yes	112542.58	No
2	3	15619304	Onio	502.0	France	Female	42	8	159660.80	3	Yes	No	113931.57	Yes
3	4	15701354	Boni	699.0	France	Female	39	1	0.00	2	No	No	93826.63	No
4	5	15737888	Mitchell	850.0	Spain	Female	43	2	125510.82	1	Yes	Yes	79084.10	No
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9991	9996	15606229	Obijiaku	771.0	France	Male	39	5	0.00	2	Yes	No	96270.64	No
9992	9997	15569892	Johnstone	516.0	France	Male	35	10	57369.61	1	Yes	Yes	101699.77	No
9993	9998	15584532	Liu	709.0	France	Female	36	7	0.00	1	No	Yes	42085.58	Yes
9994	9999	15682355	Sabbatini	772.0	Germany	Male	42	3	75075.31	2	Yes	No	92888.52	Yes
9995	10000	15628319	Walker	792.0	France	Female	28	4	130142.79	1	Yes	No	38190.78	No

Bank Churn Analysis Part-1¶

¶

Note¶

1. Please Answer all the questions.¶

2. Explain results for all the provided plots.¶

3. For any additional code, explain the code via comments.¶

Data Import¶

Content¶

Data Overview¶

Q. Why we put index_col=0 ?¶

Q. Why we copy data as df1.copy() ?¶

Q. Why we change the column names?¶

Data Cleaning¶

What do you see in the results?¶

Why we have used .astype(float) in provided code?¶

Why we are droping orginal column after using get_dummies?¶

Exploratory Data Analysis¶

Churn No-Churn Ratio¶

Varibles distribution in Churn and Non-churn category: Pie Charts for Categorical variables¶

You need to run the code and understand the output¶

Varibles distribution in Churn and Non-churn category: Histograms for Numerical variables¶

You need to run the code and understand the output¶